Select Page

Some more GNIS details

The Legal Genealogist ought to be used to it by now.

It almost doesn’t matter what the topic is — it’s pretty much guaranteed that somebody out there is going to know something or think of something I didn’t know.

And that may be the single best thing about blogging — that you find out so much you wouldn’t otherwise know.

Two cases in point, both stemming from yesterday’s post about using the Geographic Names Information System (GNIS) to find cemeteries and other features we might need to locate as genealogists.1

One is a pretty simple thing, but one I’d completely missed: the way the Geographic Names Information System lists cemeteries associated with churches. But, fortunately, fellow genealogist and friend Larry Head had run into the problem a time or two in the past, so he mentioned it in an email after reading yesterday’s post:

if a cemetery is on the grounds of a church–a churchyard–then GNIS may ignore it if one qualifies the search as a being for a cemetery. In other words, one ought to also query “church” and then check the maps to see if there’s an adjacent cemetery.


Larry’s absolutely right.

I tested this with some cemeteries I’m very familiar with and in every case, a search for the name with the feature class cemetery turned up no entries at all; searching for the name with the feature class church brought it right up. Take a look at what GNIS sees and what’s really there in Fluvanna County, Virginia, at a Methodist church cemetery in Virginia where my mother and grandparents (and a host of other relatives) are buried:

Byrd Chapel map

Only Byrd Chapel — the Methodist Church — is in GNIS. But you can see the cemetery right by it.

So… lesson number 1: remember that some feature types may be associated with other feature types and look for them all.

And, of course, that doesn’t begin to address the whole problem of what happens when the name of a feature itself has changed — when, for example, it used to be spelled one way and is now spelled another way.

That’s what happened in the case of a creek that ran into the homestead of reader Sue Richart’s husband’s family in Washington State. The ancestor was Herman Kolle, but the creek was first spelled Kohley Creek, and that’s what showed on the maps of the U.S. Forest Service.

But, she reports, “Back in the 60s, the USGS sent someone to the Stevens County Courthouse to see how Herman actually spelled his name. The board decision indicates that Herman spelled it Kolle and the USGS resolved the discrepancy and the Forest Service maps were changed to the correct name.”

Now… you’d know that in part by looking at the GNIS entry for the creek: there’s a statement right in the feature detail report that says there was a variant name Kohley Creek.2

But Sue says there’s something even better than that: “if there was a name change, you can get a copy of the decision file. When it shows a name variant and an official board decision, you can ask for a copy.” In her case, back in 2006, she had requested it by email; there’s an email contact address listed at the bottom of every GNIS page.

Duh again…

Lesson number 2: think about what might be out there in the form of those wonderful things so dear to a genealogist’s heart… records.

The things you learn when you blog… and a big tip of the hat to Larry and Sue for these particular things!


  1. See Judy G. Russell, “Where is that place?,” The Legal Genealogist, posted 3 Dec 2018 ( : accessed 4 Dec 2018).
  2. “Feature Detail Report for: Kolle Creek,” Stevens County, Washington; United States Board on Geographic Names, U.S. Geological Survey ( : accessed 4 Dec 2018).
Print Friendly, PDF & Email